EuroEval Danish NLU - ScaLA DA: leaderboard
The "ScaLA DA" subtask of EuroEval Danish NLU. Danish-language EuroEval natural-language-understanding suite, separating NLU task performance from the broader all-task EuroEval aggregate.
Metric: Matthews correlation coefficient (%). Source: euroeval.com. Saturation forecast: Estimated already saturated. 432 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3 Pro (Preview) | 77.98 | #66 |
| 2 | Gemini 3.1 Flash Lite (Preview) | 77.63 | #194 |
| 3 | Claude 3.7 Sonnet (20250219) (Thinking) | 76.38 | #201 (Claude 3.7 Sonnet (20250219)) |
| 4 | GPT-5 (2025-08-07) (High) | 75.53 | |
| 5 | O3 (2025-04-16) | 74.15 | #121 |
| 6 | Gemini 2.5 Flash Lite (Thinking) | 74.06 | #422 (Gemini 2.5 Flash Lite) |
| 7 | Gemini 3 Flash (Preview) (Non-reasoning) | 73.75 | #82 (Gemini 3 Flash (Preview)) |
| 8 | GPT-5 | 71.73 | #98 |
| 9 | Gemini 3 Flash (Preview) (Thinking) | 71.54 | #82 (Gemini 3 Flash (Preview)) |
| 10 | GPT-5.2 | 71.48 | #110 |
| 11 | GPT-5.4 Mini (2026-03-17) (High) | 71.31 | #169 (GPT-5.4 Mini (2026-03-17)) |
| 12 | Llama 3.1 405B | 70.64 | #466 |
| 13 | Claude Sonnet 4.6 | 70.3 | #90 |
| 14 | Gemini 2.5 Pro | 69.03 | #152 |
| 15 | Gemma 4 31B (IT) | 68.9 | #176 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=euroeval-danish-nlu-scala-da · How It Works · Data refreshed daily, snapshot 2026-10-20.